Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute to identify geographic variation in the use of particular treatments for cancer, and to assess whether treatment differences are associated with differential survival rates across the SEER areas. The proposed analyses will permit the assessment of whether observed geographic variation in cancer treatment and survival is independent of differences in age, ethnicity, stage of disease at diagnosis, and other selected factors that affect the receipt of treatment. The longitudinal SEER data will also allow the investigation of changes in the use of particular cancer therapies over the time period of the data (1973-1988), and associated changes in cancer survival. Finally, each of the counties served by the SEER program will be characterized with respect to variables that may affect the availability of treatment, such as proximity to a cancer treatment center, number of physicians in relation to number of residents, and population density. The cancer treatment and mortality experience of individuals from counties with easily accessible medical personnel and facilities will be compared to that of individuals from counties with fewer facilities-. Data covering the period from 1973 through 1988, and containing information on over one million individuals, are available for analysis. These analyses will provide data important to any systematic attempt to evaluate the outcomes of cancer treatment nationwide.